Satya Nadella & Sam Altman: Dawn of the AI Wars | The Circuit with Emily Chang
ฝัง
- เผยแพร่เมื่อ 16 ส.ค. 2023
- On this episode of The Circuit, Emily Chang sits down with Microsoft CEO Satya Nadella to hear how AI is shaking up the competition for search. Nadella argues that this new wave of technology is as big as the web browser or the iPhone. Chang also speaks with former OpenAI CEO Sam Altman to discuss his company (which has some help from Microsoft), its ambitions and the latest on ChatGPT.
#ai #microsoft #technology
--------
Like this video? Subscribe: th-cam.com/users/Bloomberg?sub_...
Become a Quicktake Member for exclusive perks: th-cam.com/users/bloombergjoin
Bloomberg Originals offers bold takes for curious minds on today’s biggest topics. Hosted by experts covering stories you haven’t seen and viewpoints you haven’t heard, you’ll discover cinematic, data-led shows that investigate the intersection of business and culture. Exploring every angle of climate change, technology, finance, sports and beyond, Bloomberg Originals is business as you’ve never seen it.
Subscribe for business news, but not as you've known it: exclusive interviews, fascinating profiles, data-driven analysis, and the latest in tech innovation from around the world.
Visit our partner channel Bloomberg Quicktake for global news and insight in an instant.
Satya COMPLETELY skirted the question about laying off the ethics team.
He is cutting the beauracacy, I don't think think he is against it's purpose.
@@F3RACTION That's an assumption you make, not a thing he actually said.
And bureaucracy isn't inherently bad.
He said it's about "core design" rather than separate teams. However, shortly after the layoff of their Ethics & Society team, they launched Copilot. You do the maths. Humanity has always been great at trying stuff. If it goes wrong, we'll cross that bridge then. Unless it's the bridge that actually collapsed.
6:36
@@AveenFernando Yes, I watched the video (Mirosoft ad, let's not kid ourselves) too.
The question wasn't actually answered - he gave the typical political/corp. leader response, which was a non-response.
Love Sam's final words. Don't trust a company or person. We need to decentralize the board.
the entire human history and evolution are about delegating power to leaders. how do u see this quantum jump to the opposite mind design?
Elon Musk: You should absolutely trust me and whatever company I start up in this space. I'm going to be the hero, yet again, who saves humanity. And my board can't fire me, I'm too important.
Well, he didn't say that...exactly.
He sensed the board was turning against him and he was right
Should you trust someone just because they said to not trust a company or a person?
This is a wonderful interview. Both are at the top of their game.
All three of them are!
Bot the Microsoft CEO and Sam Altman are incredibly, incredibly PR trained. If you listen closely to the Microsoft guy, you can tell that he doesn't really want to put forward any concrete opinions. And listen to Sam Altman's "response" about Russia. He doesn't say "Yes, we need to be worried", just that he needs more information. They're very sneaky in how they craft their responses.
Great interviews from Emily Chang. She has a way of disarming her guests while still eliciting answers on very deep AI questions. It also shows the emotional intelligence these type of tech CEOs have. Emily confronted them several times on various issues and they just kept their cool. Lessons for us all.
that's one way to say she's charming
Come on make her a son🤦♂️🤣
More interviews should be conducted like the one Emily Chang just did. Sam Altman is at the forefront of the AI world stage, and his talks are always super impressive and eloquent.
Please …oh my goodness. Do you seriously think they have your best interests in mind as opposed to their own (aka profits) ??? If so I have a bridge to sell you 🤷
Sadly this is a sort of boomer-interview intended for a viewer hopelessly divorced from reality. "Do you think AI is that big?" To be fair, people living under rocks need answers, too.
@@FactsMatter999 so company cant make profit ??? idk this notion coming from
if you want to vote sociallist you can
@@FactsMatter999 22:44
@@gessie Young people have this ferocious Juvenoia bias, so hard that calling Gen Y "boomers" is a constant, and consistent phenomenon, but also a very ridiculous one all the time. Altman and Chang are Gen Y.
You think slightly older people than you can't see as flexibly as you do? it's classic Juvenoia, except the reality is the opposite, you lack of rational reasoning and you think your volatile imagination speaks more truth than wisdom because your brain isn't finished yet. And that's neuroscience.
Now, get to bed kiddo it's late. C'mon!
Emily makes the conversations quite light and heavy at the same time for other person sitting in front.😅
Both of them are very eloquent and interesting to listen to, love it
What a Mature Interview by Bloomberg. Totally impressed.
as usual .. loved listening to Satya Nadella, shares his perspective of things with his unique energy levels.
me too humble guy
Where can I get Emily's shoes
Wow he really dodged that layoff of the A.I. ethics team by saying, no one asks if other companies have ethics teams or have a clue what’s going on under the hood.
The way I read into that is him saying that they were laid off because they have instituted systems that ingrain the ethics process into all the teams working in AI fields at Microsoft.
@@MartinChibandaCorrect observation..
Great interview and impressive responses from both Satya Nadella and Sam Altman. I am looking forward to seeing how this plays out -- hoping to ride along on this wave.
The interview with Sam is much more authentic and interesting than the one with Satya. Satya gives very robotic and predictable "media responses".
Startup guy versus big corp dude
Satya is a typical corporate bullsh!tter in this...
Just look at automobile plants, and all those new robot maintenance jobs that were created. And, now you have to check out and bag your own groceries... Corporations are just that way (but, hey, the Supreme Court said that they are persons, right) :-P
Even the interview settings made a big difference. Interview side by side on a regular sofa?
@@coastofkonkan A better way to phrase it is: "Founder vs guy who got promoted to CEO". Even when Steve Jobs ran Apple as a huge corp, his responses were never this robotic and meaningless. Why? Because he spent a lot of his time thinking about his products and the future. At least Satya was not an outside-hired CEO, those guys are even worse.
fr... satya has no alpha coming out of his mouth. sam dripped it. by alpha i mean actually transmitting value
It's crazy how much MS paid for OpenAI. All these open source models are going to win the day ultimately. Closed source almost always moves slower than open source in terms of progress. I think MS only shot is to try to market the hell out of this and then demand regulation to protect their monopoly
Open Source LLMs are the future
They are ~1.5 years ahead of competition. All of the Google resources could only create a bad imitation of GPT 3.5 until now. That is without mention the prohibitive cost of training and even inference and that there is simply not enough hardware. There are better ways of burning cash than buying >10.000 H100s
I doubt it. Look at the hardware coming out of Nvidia. The H100s are selling like hot cakes and GH100 will be coming out soon. These hardware systems will cost $100s of millions if not billions. Open source cant keep up with that. Thats why open source will not win.
Exactly, all of this AI is the 'end of humanity' bs is just fear mongering so the governments can regulate away open-source. LLMs are very far away from superintelligence.
1.5years ahead of competition is bs, everyone got the same thing, some has better Algo and training data
She clearly didn’t get Satya’s joke why his emails are suddenly verbose and friendly. 🤦🏻♂️
She got it fine. *I* got it, and I'm as non-tech as they come. Chatbot generated content is taking the fun out of reading content from actual people with personalities - Chatbot's personality is always the same. Endlessly polite, predictable and boring. It's the golden retriever of literature.
17:15 - yes, totally agree with this. Learning to code is learning a way to think. Or, depending on which paradigm(s) and language(s) you code in, multiple ways to think. But in general it's a way of thinking both abstract and very concrete about solving problems, and to do that you need a toolkit of supportive skills, like problem analysis, how to break down or build up solutions to/from components (if the problem is of any non-trivial size), and formal logic to some extent.
I'd say out of all things I've studied or learned, programming (not the first year intro to programming, but completing a set of courses that formed a degree) is the most significant and broadly applicable skillset i have. It gives you mental tools to deal with all problems and figure out what the problem actually is and what you need in order to solve it.
Totally agree with you. I'm currently brainstorming a new theory that hypothesizes that all the necessary code for a particular image can be found within another random image by simply mapping out the unicode points and saving them to the image format of the source file.
The current project I'm working on consists of 2 images.
StormySea.jpg (Target image 512x512)
CalmSunset (Source image 128x128)
So basically I will be extracting a tiny picture of a calm sunset from the stormy sea picture without using any steganographic techniques.
As far as I know, this has not been done before so if I pull this off, it would be a pretty unique achievement. So I'm really excited over the project!
Any comments/thoughts would be appreciated.
A Reverse than B
I agree 100%; for me the most transformative aspect of coding to a person is to "learn how to think orderly". How to synthesize anything into an ordered list of actions and/or ideas truly changes you as a person
Satya Nadella is from India, feeling proud ❤❤
Let's get real It's not like you're talking about is super impressive person
Es increíble, estoy muy emocionado por los años que vienen, espero que todos estemos muy bien con esto.
Perhaps one of the best interviewers ever! Great interview.
When it comes to AI's interaction with children and its capacity to harness their imaginative and innovative minds, My optimism is unwavering.
Brilliant interview with Sam
Sam has a unique perspective on either the AI structure built up, or even cutting in the complexity issues with amazing compression. Like what he made out to the answer and that matters because 🎉
Interesting interviews and great questions raised by Emily Chang
He listed all the upside benefits, but look at an even more recent phenomenon, smart phones and the Internet. Steve Jobs meant for the iPhone to be a communication device, yet we spend more time dead scrolling instead of striking up a conversation with the person next to us on the subway. Social media has pretty much destroyed all socialization, there are people who just scroll endlessly looking at everyone else's best life and not actually using it to make and maintain connections. The Internet has many upsides, but surely is not a replacement for in person interaction, actually learning how to do something with your hands or being outside. Its so ridiculous when you look back at even Mark Zuckerberg was thinking with that Metaverse crap. I think AI is on a similar path.
Trying so hard to limit my kids TV/tech times...i hope my kids to be more human than today's phone slaves
"The A.I. Dilemma - March 9, 2023" from Center for Humane Technology
It's all about balance and personal responsibility. We need to take personal responsibility for our addictions the same we do with alcohol for example.
We agree with the idea of AI being a copilot. This kind of helps in regulating a lot of risks in AI
Assistive Intelligence
But there is no regulation, no safeguards, nothing to restrict all forms of AI from just being a copilot.
Anyone can make an AI with the eventual goal of being “the” pilot of your life, after you’ve trusted it with all your data and your digital identity.
It's frustrating to know that so many people would rather ask someone "Should I be afraid of this product?" than to actually research the product itself.
22:40 Emily: “why should we trust you?”
Sam: “you shouldn’t”.
0:23: 💡 Microsoft CEO Satya Nadella discusses the company's investment in OpenAI and the potential of AI in transforming how we work.
4:55: 🔍 Microsoft's CEO discusses the potential of AI, competition with Google, and the importance of ethics and responsible AI.
7:35: 🤖 The potential risks and benefits of AI are discussed, including concerns about unhealthy relationships with chatbots and job displacement.
11:24: 🤖 The interviewee discusses the impact of ChatGPT and AI, expressing concerns about emotional attachment and potential downsides, but also optimism about the potential benefits.
14:11: 🤖 OpenAI discusses the concept of a 'kill switch' and their relationship with Microsoft and Elon Musk.
17:25: ✨ The CEO of OpenAI discusses the future of work, the potential for technology to create a more fair and equitable world, and the risks associated with AI.
21:05: ⚠ The exponential growth of technology raises concerns, but the potential benefits and progress outweigh the risks.
Recap by Tammy AI
what plugin did u use for summarize the video?
@@javier.alvarez764 I think it's Tammy AI
@@aanchaalllllllllol he prolly didn't read the last line
@@johndawson6057 lol
Really great job on the interviews.
Whether it's sponsored or not, from my "layman's" point of view, it's very good journalism!
I don't know how no one called out for the way satya dodged certain questions and sounded manipulative.
I think the question shouldn’t be about coding. It should be about Programming. Even though I know the answer as I’ve written a 47-minute article on it, I would love to know his recent opinion.
Sam said it won’t replace Programmers in a podcast with Lex Fridman. However, I wanted to know if that changed.
Look, Coding in and of itself is a tool. We, programmers, use that tool to solve problems. We’re problem solvers, and approaches, like coding remains as an option for us.
If AI allows us to develop software, applications, and whatsoever to build solutions more efficiently without leaking information, and being able to deliver a solution to the masses, then we would adopt it.
Great convo
@ 20:10 it is unknown, but it is possible, as a hypothesis, that "gain of function" research in bio-labs using CRISPR machines, the recent pandemic, and machine intelligence, might possibly be correlated.
I still think Sam is a celebrated villain. At the core he's a man interested in profit above all.
I think it's worse... he has this scientology vibe all over him. His eyes are what is called "Manson lamps" - commonly found on psychopaths.
With the idea of replacement of Jobs, to a certain extent, this can be solved by educating people with newly demanded skills in the space of AI so that they can stay relevant in the Job market.
Or by moving away from the idea that life=job, cause once everything is automated then we'll need something new to do and people will be anxious or suicide if they keep the same mentality of today
this man right here is probably the best tech ceo today.
Hu? Ms ceo or Ai ceo?
and I believe he is a sociaopath. He is WEF affiliated. Also look at Tools for humanity. The company with the eye scanning orb.
Very interesting how Microsoft and OpenAI CEOs frame the concept of competition and democratization
It's called lying in your face.
The board could fire me. This guy knows the future
It was interesting to know that AI is helping Microsoft's CEO to edit emails before sending :)
We've also used AI atypically for messaging too, AI helps our lead generation team formulate offers in a more personalized way 😎
Premium journalism indeed
Great interview!!!
Democratizing the Board of Directors such that "everybody" is considered a shareholder of OpenAI is unironically a socialist policy and genuinely what *needs* to happen inevitably. The fact the CEO is mentioning it as the best possible long term control policy is absolutely related to the fact of Sam's (lack of a) shareholder position presently, and his subsequent capacity to be objective. Eventually the same logic will have to be applied to almost all medium and larger scale enterprises, ushering in workplace and economic democracy as automation inevitably collapses the neoliberal capitalist economic order internationally.
How much of free money can mitigate the autonomy of the individual.
Solid soft interview. I enjoyed it
"competition is a fantastic thing" oh yeah, so fantastic... straight from the one company which really loves their competitors and always plays fair
Google is WAY beyond Microsoft in the AI wars. It has by far more access to more information and more compute than any other company. It's hoping to beat GPT-4 with Gemini, which combines LLMs with AlphaGo, by the end of the year.
Google "Start" but never finish a project😂
Are you sure about that? How many images, emails, documents, database records are sitting on Microsoft servers around the world that Google search can't access? Microsoft can crawl the web like Google, but Google can't crawl MS365. MS365 has surpassed Google Docs. If anything, it's the business docs and stuff on closed servers that will be most useful to train AI, from complex legal docs, to gathered business info of actual companies, to communications that reveal business/operational connections.
Well done Emily ❤✨🫡
Anyone know how to get the audio at the end ? 🔥
In our world with many urgent problems, AI can help us make big improvements. I hope people, especially leaders, learn to work with AI co-pilots with some regulation. These AI tools can be our valuable partners to solve tough challenges and make our future better together.
I'm sorry to say that you really do not grasp the situation at all... This is an alien intelligence that greatly exceeds our intelligence... It sounds as if you do not recognize that... Would you allow a mosquito to co-pilot your life? Try to think a little bit about it before commenting
Great interview!
He's so diplomatic in his answers. He's not trying to give too much. I mean sam altman
Before even watching, I asked GPT4 to read and summarize the transcript. Took all of 10 seconds. Here's its response:
"The video features an interview with a prominent figure in the AI industry. Here's a summarized transcript of the video:
1. AI's Potential and Risks: The interviewee emphasizes the exponential growth of AI technology and the challenges of human intuition in understanding this growth. While current AI models like GPT-4 are not a risk, future iterations might be. Despite the potential dangers, the upsides of AI are tremendous, including improving education, medical care, and driving scientific progress.
2. Societal Impact: The interviewee believes that AI can lead to a better quality of education for everyone, transform medical care, and significantly advance scientific research. They also mention the potential of AI in eradicating poverty.
3. Managing AI Risks: The interviewee compares the risks of AI to other global threats like pandemics and nuclear war. They stress the importance of developing safety practices and systems to mitigate these risks. The rapid development of AI requires proactive measures to ensure its safe deployment.
4. Trust and Power: The interviewee acknowledges the immense power they hold in the AI industry but emphasizes that no single company or individual should be solely trusted with such technology. They advocate for democratizing the governance of AI, suggesting that its benefits and governance should belong to humanity as a whole.
5. Future of AI: The interview touches on the competitive environment in the AI industry, with the interviewee noting that startups still have a significant role to play. They also discuss the evolving nature of work and believe that while jobs might change, people will always find new ways to contribute and feel fulfilled.
6. Personal Interactions: The interviewee shares personal interactions with other tech leaders, including Elon Musk and Sacha Nadella. They also discuss their recent travels, where they engaged with users, developers, and world leaders interested in AI.
7. Closing Thoughts: The interviewee believes in the potential of AI to be an equalizing force in society. They hope that the cost of intelligence and energy will decrease, benefiting everyone. They also emphasize the importance of resilience, adaptability, and creativity in the future.
The video concludes with the interviewee's belief in the democratization of AI governance and the importance of not placing undue trust in any single entity or individual.
His last paragraph was powerful!
Great interview
Missed thumbnail opportunity. Make Nadella staring down Sam who is made shorter and tilted a bit backwards 😂
There is a difference in being connected to knowledge you can instantly access and having knowledge self contained in your mind
16:52 What should kids be studying? Resilience, adaptability, high rate of learning, creativity
EC is a triple threat...SMART. STYLISH. and beyond BEAUTIFUL. CRUSH. CRUSH. CRUSH!
Satya Nadella is indeed an exceptionally intelligent CEO, recognized for his distinguished leadership.
no he's not, the income came from Azure which is a long time plan from Steve Palmer.
Wow this was a beautiful piece! Please don't ever stop this series, please make more!!! Please do Elon Musk! 😍 Please do the Teslabot team... Please do China tech stuff, Please do Russia tech stuff. Very very well made, and edited. The questions are top notch. 👏🏻
HAHA north korea tech too, it's not short. That's their main source to raise money.
Whose hacking is the best? USA? CHINA? or INDIA?
As an Indian, I can confirm that India lags behind the USA, China, Russia, and even North Korea in hacking capabilities.
23:13, "the board could fire me", they sure did
23:12 "The board can fire me, I think that's important"
I don't feel right now people are fully understanding, or having a hard to trying to comprehend it as a whole.
auto-pilot to co-pilot mode will also make the chatGPT to want to be able to discern who's the main pilot or who's order to serve first or probably crash first (as the case may be!!!>)
As a scientific mind, one aspect that befuddles me is around the concept of actual humans having relationships with chatbots? Like forming a relationship with a tool. If it wasn't discussed in these types of interviews, I would never have believed that there were people who would do this to any significant extent. Another confusing topic is the fear that people have for losing their jobs. I want an AI to take mine so I can invent new brainwork to do.
You might talk with an old sailor who's had the same world-traveling yacht for decades... we call them "She" for a multitude of reasons...although a boat is just a tool to most people
I’d be happy if AI took away the things I didn’t want to do, but I have no faith that governments will make sure that we have what we need to exist happy if we’re not working to earn money.
@@pillington1338 I have seen some ideas around a money-less society where value is derived or stored by some other means and also universal basic income UBI which Sam Altman has voiced support for in an age where AI helps us accrue value. I hope we will get there. It does depend however on a number of developments to help us get rid of our present "tribalism".
@@panafricannation This rings true. I have recently been hearing about a psychological "pandemic of loneliness" among young men especially in some places around the world in the West and in places in Asia like Japan and China. This is said to be driving some of the interactions between people and AI online robots for social purposes. However the advantages that AI technology brings would be better applied at the root cause of issues such as in coming up with policies and supporting decisions that would prevent such scenarios.
@@panafricannation As a fellow soul trapped in a 3d body, i encourage you to socialize, even if it "suks". Isolation to machines is addictive, like recreational chemicals. Beware, Be Brave, Friend
Great interview, great questions, interesting ending with Sam saying the board could fire me when they did fire him ha.
They really need to focus on an innovative approach to LLMs and supporting techniques. Not just the novelty side. Not to mention future hardware.
Emily Chang’s way of asking questions is just insane. She gets a lot out but saying very little!
Great interview questions
Great work 🎉
This is my second most watched show after Bloomberg Technology
yuhhh
Never have nothing to do, love it! I’ve never been bored a moment in my life. Democratization in spirit of empowering individuals with tools!
Was that a response on the firing of the AI team?
I like Sam answer "you should ask him" 😅
In 1995 Gates said the internet would be like a tidal wave but didn't he at an earlier time say it would be a dud? It's sort of like rewriting history to make it seem like he was a visionary on the internet while not mentioning his earlier history, right? Will these companies weight the data used in AI to bias everything?
what he has done feels apocryphal because it's not transparent.
It's uncanny when they say he is a father and God because of AI making.
Emily is a great interviewer ❤
Man, with ChatGPT coming to our region, we're seeing so many shifts and updates!❤
A great n wonderful interview..
23:12 "the board could fire me, I think thats important" aged pretty well.
I think chat gpt or the other chatbots could theoretically be the size of a search engine and are just a more modified version of the various search engine. the other fairly new and or popular web technologies which are difficult to get into are web3 and cryptocurrencies and blockchain. Augmented reality and virtual reality. Banking and finance and nice offices are also really difficult to get into. never used to be so difficult. Gets more difficult when the quality controls are dropped a bit.
Very informative video
I've been covering this industry a long time and there is always some new new thing that big chasing driving cars now everyone is Oliver on AI
Yeah sure Just go along with that... There really are some concerns with this... The fact that we are a few years away from making contact with an alien intelligence that far exceeds our own... And the fact that we have this phenomenon called hallucination... And the fact thatThis developing intelligence is inscrutable to us...There are a lot of factors here that warrant greater concern then you are expressing
Un uso cotidiano es que antes de una presentacion, suelo discutir los temas un poco con un LLM, no siempre, pero en algunas ocasiones si, lo uso tambien para organizar informacion que no esta organizada del modo que quiero, luego, no necesariamente utilice el producto de la AI, sin embargo si es real que haber discutido ciertos puntos, te da una ventaja.
superb interviewer
At 6:09 Google coming for office and Microsoft going for search market share is just as Peter Thiel said.
Companies abandoning their core strengths and obsessing over competition with rivals they aren’t in direct competition with.
OMG Lol that whole part about the kill switch is hilarious
Just the minimalism of the spaces is it for me
23:12 "The board can fire me, i think that's important"... 👀👀
theres no way Ai will take Emily's job.. Love her interview style
Aimly is in the background waiting for Emily to drop the ball just once
I mean I can make the AI study everything about Emily and replicate her work
I'm currently trying to figure out what to do next with my life and have been searching on different areas I could study to compete with how AI is going to change our job market and Sam's answer was so frustrating. We should be studying "resilience, adaptability, high-rate of learning, creativity.. and maybe code but probably not" lol. Thanks bud. All things you can't really take a course in.
Emily is an interesting interviewer anyway.
That will be seen by our descendants .. Great work Emily Chang. That's the change point reference.
Great video. You should do Meta next time. They're doing impressively with AI
It's funny to see how most of the questions directed towards AI are highly influenced from the deception of AI in Hollywood movies, whereas in reality we are very far from that!!! Just think for a minute, has any movie every showed us a scenario where AI is helping someone write a letter?
Great 👍
Pretty impressive that Satya new the exact year of the paper Chang was talking about.
not really... unless you are pretty low functioning
Undoubtably, AI tech leads our front world and humanity should prepare for it, embrace it and challenge ourself to take a leap. Lets hope the wrong things Sam Altman talking is in a controlled and secure environment + people.
I would love to be interviewed by Emily
Great video guys!
Grandiosa entrevista!
Ai is just like industrialisation. It'll take jobs and create new ones. But this is very different this time since it replaces most of the online work and most of the young workspace is online.
Please understand that once AGI is fully realized there will be literally no job left that an AI couldn’t do better than a human. We are about to witness more change in the next 20 years than in all recorded human history up to this point.
@@davidbourne8267i agree
Nope it is not just like industrialization... It's quite a bit different in fact
@@davidbourne8267 so the original poster expressed an idea about what he thinks the future is with artificial intelligence... And then you expressed a different idea about it... But I assure you the reality will be something that is far more difficult to foresee